Spatiotemporal Data Cube Modeling for Integrated Analysis of Multi-Source Sensing Data

One of the significant challenges of spatiotemporal big data management is the inconsistent data structures and organization methods of different data sources, such as remote sensing raster data and location-aware social sensing data. Traditional spatial data management systems are unable to support the integrated process and analysis of those data with different data types. Currently, in the field of remote sensing data management and processing, the data cube model is promising for constructing "analysis ready" remote sensing big data. This paper is initiated by the data cube model from data warehousing research area, and investigates the integrated data model and organization methods for remote sensing data and location-based social sensing data based on spatiotemporal data cube. Moreover, we implement the spatiotemporal data cube model on an array DBMS, and show a disaster analysis scenario for integrated analysis of multi-source spatiotemporal big data.